Strengths and Limitations
A major strength of our study is the large sample size including 43,721 fetuses with low-pass GS and ultrasonography. Most previous studies explored the yield of pCNV in fetuses with ultrasonographic phenotype by using CMA frequently in small cohorts. Secondly, all anatomical system anomalies were included rather than several common systems only, which makes our study more representative and generalizable. In addition, the contribution of aneuploidy/pCNV to fetuses with ultrasonographic abnormalities was evaluated by comparing with the group without identifiable anomalies, providing a powerful and objective guidance for prenatal genetic diagnosis.
This study had some limitations. There was no further ultrasonographic imaging
follow-up, which might have overestimated the contribution of chromosomal aberrations in fetuses with normal ultrasonographic findings. Also, postnatal outcome
data were unavailable to verify prenatal ultrasonographic anomalies. Furthermore, this study only considered chromosomal-level variations, and lacked gene-level variations. We expected future study could combine whole exome sequencing and cytogenetic methods to improve the identification of genetic disorder in fetuses with ultrasonographic anomalies.34-36 Meanwhile, due to lack of fetuses with normal ultrasonography and maternal age <35 years, no comparison was performed for the yield of aneuploidy between fetuses with abnormal ultrasonographic findings and control group. Finally, some overlap may occur between soft markers and anomaly groups, as reports of increased nuchal translucency, thickened nuchal fold and cystic hygroma from different centers and at different gestational weeks, may present within the same group. Despite our best efforts to classify the ultrasonographic anomalies, an international uniform classification system is still lacking. Therefore, we advocate implementation of such a system to facilitate the comparability of cohorts.